R Notebook

library(plotly)
source("/home/and/Documents/PhD/Research/Scripts/R/R13/tasksMetrics.R")

appsE1_1 = "/home/and/Documents/PhD/Research/Scripts/Misc/tmp/apps/app-0282_info.tsv"
appsE1_2 = "/home/and/Documents/PhD/Research/Scripts/Misc/tmp/apps/app-0284_info.tsv"
appsE1_3 = "/home/and/Documents/PhD/Research/Scripts/Misc/tmp/apps/app-0286_info.tsv"
stagesE1 = rbind(getStages(appsE1_1),getStages(appsE1_2),getStages(appsE1_3))
appsE3_1 = "/home/and/Documents/PhD/Research/Scripts/Misc/tmp/apps/app-0281_info.tsv"
appsE3_2 = "/home/and/Documents/PhD/Research/Scripts/Misc/tmp/apps/app-0283_info.tsv"
appsE3_3 = "/home/and/Documents/PhD/Research/Scripts/Misc/tmp/apps/app-0285_info.tsv"
stagesE3 = rbind(getStages(appsE3_1),getStages(appsE3_2),getStages(appsE3_3))
stages = rbind(stagesE1, stagesE3) %>% select(StageId, Stage, Executors, Duration) %>%
  mutate(Stage = paste(str_pad(StageId,5,"left"), Stage)) %>%
  select(Stage, Executors, Duration) %>%
  group_by(Stage, Executors) %>% summarise(Duration = mean(Duration))
head(stages)

p = ggplot(data = stages, aes(x = Stage, y = Duration, fill = Executors)) +
  geom_bar(stat="identity", position=position_dodge(width = 0.75), width = 0.7) + 
  theme(axis.text.x = element_text(angle = 90, hjust = 1)) +
  labs(x="Stages", y="Duration(s)")
ggplotly(p)